Don't Read the Log: Execution Traces Contaminate Verifiers in Video-Generation Agents

arXiv:2609.28564 · cs.CR, cs.LG · Submitted 2026-09-23 · Read on arXiv

cs.CR, cs.LG

Submitted: 2026-09-23

Updated: 2026-09-23

Code: https://github.com/JaidedAI/EasyOCR

Terminology

• Trace — 51× in this paper · explained in 2 episodes
• Verifier — 11× in this paper · explained in 2 episodes
• Accuracy — 7× in this paper · explained in 1 episode
• Benchmark — 7× in this paper · explained in 2 episodes
• Verification — 5× in this paper · explained in 2 episodes
• Consistency — 4× in this paper · explained in 3 episodes
• Hallucination — 3× in this paper · explained in 4 episodes
• Benchmarking — 2× in this paper · explained in 4 episodes
• Generative AI — 2× in this paper · explained in 2 episodes
• Model-Agnostic — 2× in this paper · explained in 2 episodes
• Reward Hacking — 2× in this paper · explained in 8 episodes
• Vision-language models — 2× in this paper · explained in 2 episodes
• Actor — 1× in this paper · explained in 3 episodes
• Adversarial Robustness — 1× in this paper · explained in 3 episodes
• Agent Harness — 1× in this paper · explained in 2 episodes
• Agent-as-Judge — 1× in this paper · explained in 1 episode
• Calibration — 1× in this paper · explained in 7 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Generative Models — 1× in this paper · explained in 3 episodes
• Grounding — 1× in this paper · explained in 3 episodes
• LLM judge — 1× in this paper · explained in 2 episodes
• Multi-agent Collaboration — 1× in this paper · explained in 2 episodes
• Provenance — 1× in this paper · explained in 4 episodes
• Robustness — 1× in this paper · explained in 5 episodes

Sources

• Let's Think in Two Steps: Mitigating Agreement Bias in MLLMs with Self-Grounded Verification
• Qwen3-VL Technical Report
• Qwen2.5-VL Technical Report
• DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
• OmniHalluc-L: Counterfactual Benchmarking and Modality-Perturbation Reliability Calibration for Long-Form Omni Hallucination
• Evidence-RL: Towards Evidence-intensive Visual Reasoning
• ViMax: Agentic Video Generation
• EG-VQA: Benchmarking Verifiable Video Question Answering with Grounded Temporal Evidence
• Seeing Isn't Believing: Uncovering Blind Spots in Evaluator Vision-Language Models
• The Curse of Multi-Modalities: Evaluating Hallucinations of Large Multimodal Models across Language, Visual, and Audio
• SoliReward: Mitigating Susceptibility to Reward Hacking and Annotation Noise in Video Generation Reward Models
• UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild
• Wan: Open and Advanced Large-Scale Video Generative Models
• On the Adversarial Robustness of Multimodal LLM Judges
• VideoWeaver: Evaluating and Evolving Skills for Agentic Long Video Generation
• When Does a Video-Language Model Stop Watching? Reward Strength Controls the Formation and Reversal of Visual Shortcuts in Multimodal RLVR
• SPIRAL: Self-Evolving Action-Conditioned Video Generation via Reflective Planning Agents
• VideoAgent: All-in-One Framework for Video Understanding and Editing
• When Vision-Language Models Judge Without Seeing: Exposing Informativeness Bias

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills
• Token Composition: A Graph Based on EVM Logs